A non-catalytic role of RecBCD in homology directed gap repair and translesion synthesis

Abstract

The RecBCD complex is a key factor in DNA metabolism. This protein complex harbors a processive nuclease and two helicases activities that give it the ability to process duplex DNA ends. These enzymatic activities make RecBCD a major player in double strand break repair, conjugational recombination and degradation of linear DNA. In this work, we unravel a new role of the RecBCD complex in the processing of DNA single-strand gaps that are generated at DNA replication-blocking lesions. We show that independently of its nuclease or helicase activities, the entire RecBCD complex is required for recombinational repair of the gap and efficient translesion synthesis. Since none of the catalytic functions of RecBCD are required for those processes, we surmise that the complex acts as a structural element that stabilizes the blocked replication fork, allowing efficient DNA damage tolerance.

Figure 1.

(A) Model of the replication fork encountering a lesion in the leading or lagging strand. (B) Genetic system to monitor sister-strand exchange mechanisms (modified from (4)). The scheme represents the situation in which the lesion (red triangle) is located in the 5΄-end of the lacZ gene in the leading strand. The damaged strand containing the marker D, where the lesion is located, and the marker C, placed 100 bp upstream the lesion, contains a +2 frameshift in order to inactivate the lacZ gene. Opposite to the lesion we introduced a +4 loop (marker B) that restores the reading frame of lacZ, and in the same strand we added marker A that contains a stop codon. Therefore, the two strands are lacZ-. Only a mechanism of HDGR by which replication has been initiated on the damaged strand (incorporation of marker C), and where a template switch occurred at the lesion site (leading to incorporation of marker B) will restore a lacZ+ gene (the combination of markers C and B contains neither a stop codon nor a frameshift). Using the same system we can also score for TLS events (combination of marker C and D), as sectored pale blue/white colonies, and for damaged chromatid loss events (combination of marker A + B), as pure white colonies. *For the combination of marker C and D we observed a leaky activity of the β-galactosidase due to a translational frameshift.

Figure 2.

Partitioning of DDT pathways in the recB nuclease deficient strains (recB^D1080A). The graph represents the partition of DDT pathways in the presence of the UV lesion TT(6-4) inserted in different recombinant deficient strains. The lesion has been inserted in either the leading (lead) or the lagging (lag) strands of E. coli chromosome. Tolerance events (Y axis) represent the percentage of cells able to survive in presence of the integrated lesion compared to the lesion-free control. rec+ corresponds to our parental strain, recombination proficient. The data for rec+, recF-, recB- and recF-recB- strains have been previously published (4). recB^D1080A strains deficient for the nuclease activity are indicated in red. The data represent the average and standard deviation of at least three independent experiments. T-test was performed to compare values from the different mutants to the rec+ strain, with the exception of recF-recB^D1080A whose values have been compared to the single recF mutant. We also compare values for the leading and lagging orientation of the recB and recF recB mutant strains. For HDGR: *P < 0.05; **P < 0.005; ***P < 0.0005. For Damaged chromatid loss: •P < 0.05; ••P < 0.005; •••P < 0.0005. For survival: ⁺P < 0.05; ⁺⁺P < 0.005.

Figure 3.

Partitioning of DDT pathways in recD mutant strains. The graph represents the partition of DDT pathways in the presence of the UV lesion TT(6-4) inserted in the recD- and the recD^K177Q mutant strains (indicated in red). The lesion has been inserted in either the leading (lead) or the lagging (lag) strands of E. coli chromosome. Tolerance events (Y axis) represent the percentage of cells able to survive in presence of the integrated lesion compared to the lesion-free control. rec+ corresponds to our parental strain, recombination proficient. The data for rec+ and recB- strains have been previously published (4). The data represent the average and standard deviation of at least three independent experiments. T-test was performed to compare values from the different mutants to the rec+ strain. We also compare values for the leading and lagging orientation of the recB and recD mutant. For HDGR: *P < 0.05; **P < 0.005; ***P < 0.0005. For Damaged chromatid loss: •P < 0.05; ••P < 0.005; •••P < 0.0005. For survival: ⁺P < 0.05; ⁺⁺P < 0.005.

Figure 4.

Partitioning of DDT pathways in strains expressing the recA730 allele. The graph represents the partition of DDT pathways in the presence of the UV lesion TT(6-4) in strains expressing the recA730 allele or the wild-type copy of RecA (indicated in red). In those strains the sulA gene has been also inactivated to avoid cell division blockage because of the constitutively SOS activation due to the recA730 allele. The plasmid pLL59 contains the recA730 allele, while the plasmid pLL58 contains the wild-type copy of recA gene. The lesion has been inserted in either the leading (lead) or the lagging (lag) strands of E. coli chromosome. Tolerance events (Y axis) represent the percentage of cells able to survive in presence of the integrated lesion compared to the lesion-free control. rec+ corresponds to our parental strain, recombination proficient. The data for rec+, recF− and recB− strains have been previously published (4). The data represent the average and standard deviation of at least three independent experiments. t-test was performed to compare values from the double mutants (recA recF or recA recB) to the their correspondent single mutant strain (recF or recB). For HDGR: *P < 0.05; **P < 0.005. For damaged chromatid loss: •P < 0.05; ••P < 0.005.

Figure 5.

RecBCD complex is involved in TLS pathway. The graph represents the percentage of TLS events in the presence of three different replication blocking lesions (TT6-4, G-AAF, dG-BaP(−)) in the recB-, recB^D1080A and recD^K177Q strains (indicated in red) in comparison with a rec+ strain. The data for rec+ strain have been previously published (41,49). The data represent the average and standard deviation of at least three independent experiments in which the lesion has been inserted either in the leading or the lagging strand. The data from leading and lagging strands have been pooled together because no significant difference was observed between the two orientations. t-test was performed to compare TLS values from the recB- strain to the rec+ strain. *P < 0.05; **P < 0.005; ***P < 0.0005.